• 
• 

Three Part Question

In [critically ill patients with acute renal failure] does [the use of loop diuretics] [reduce mortality, improve renal function, reduce length of ITU/hospital stay or reduce requirements for renal replacement therapy]?

Clinical Scenario

A 65 year old male presents to the emergency department with a severe pneumonia. He is intubated and placed on a ventilator because of worsening hypoxia. He has no history of previous renal disease however he becomes increasingly oliguric over the next 2 hours despite adequate fluid resuscitation and vasopressor support. You wonder whether the administration of a loop diuretic in order to improve/maintain his urine output will improve his prognosis and reduce the need for continuous veno-venous haemofiltration (CVVH).

Search Strategy

Cochrane Database - Edition 2 2006
OVID Medline 1966 to February Week 1 2006.
Embase 1980-Feb 2006.
(exp Uraemia OR azotemia.mp OR exp Kidney Failure/ OR exp kidney failure, acute/ OR exp kidney tubular failure, acute/ OR acute tubular necrosis.mp OR acute renal failure.mp OR renal failure.mp OR acute kidney failure.mp OR ARF.mp) AND (exp Diuretics/ OR Diuretics.mp, OR loop diuretic$.mp OR exp furosemide/ OR frusemide.mp OR exp bumetanide/ OR bumetanide.mp OR burinex.mp OR lasix.mp OR torasemide.mp OR Torem.mp) and ((critically ill or critical illness).mp OR exp Critical Illness/ OR critical care.mp OR exp Critical Care/ OR intensive care.mp OR intensive care units.mp OR exp intensive care/) limited to English and humans.
Cochrane: loop diuretics

Search Outcome

97 papers were found on Medline and 721 papers on Embase of which only 2 directly answered the three part question.
There were 209 citations in Cochrane. No new papers were found.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Uchino et al, 2004, Australia (Multicentre study 23 countries)	1743 patients on intensive care units on renal replacement therapy or with acute renal failure (using pre-defined criteria). 70% of patients were treated with diuretics at study inclusion. Frusemide was most commonly used diuretic (98%).	Prospective multicentre (54 centres), multinational (23 countries) cohort study. Three distinct multivariate analyses were performed using propensity scoring.	Unadjusted hospital mortality rates	Diuretic group (62.4%) v controls (57.1%). Odds ratio mortality 1.25 (p<0.03).	Observational cohort. Heterogeneity of care across 23 countries. Analytical methods can only adjust for observed confounding variables not unobserved ones.
			Adjustments using three multivariant models	Overall Odds ratio mortality similar for all three methods. 1.21-1.22 (CIs 0.92 - 1.6) p=0.10 to p=0.153.
Mehta et al, 2004, USA	552 critically ill patients with acute renal failure in 4 ITUs in California. 326 patients (59%) were treated with diuretics.	Prospective cohort study October 1989-1995.	Relationship between diuretic use and mortality and non-recovery of renal function	Increased risk of death and non-recovery of renal function with diuretic use. Odds Ratio 1.77 (CIs 1.14-2.76) in group given diuretics.	Prospective observational cohort study. Patients given diuretics were likely to be older and more likely to have a history of CCF (p<0.001) and had lower measured Cardiac Index (p<0.001). Some inclusion criteria debatable.
			Odds ratio of death when patients who died in week 1 were excluded	Odds ratio of death 3.12 (CIs 1.73-5.62)
			Overall in-hospital mortality	Odds Ratio 1.68 (CIs 1.06-2.64)

Comment(s)

About one fifth of the cardiac output is directed to the kidneys. This exceeds the oxygen supply to other vital organs, such as the brain, heart, or liver. A very low fraction of oxygen delivered is extracted by the kidney, suggesting ample oxygen reserve. Paradoxically, the kidney is the organ which is most sensitive to hypoperfusion and hypoxia, with acute renal failure being one of the most frequent complications of hypotension. This is because of the physiological gradient of intrarenal oxygenation, which means that under normal physiological conditions the renal medulla functions at very low oxygen tensions. Many therapeutic interventions in the prevention or management of patients with acute renal failure have been investigated in clinical studies. Interventions to enhance renal blood flow and decrease tubular reabsorption seem to be a logical approach for the prevention of outer medullary hypoxic injury. Loop diuretics block the active sodium-potassium-chloride co-transport in the apical membrane of the thick ascending limb renal tubular cells. The loop diuretic frusemide has been shown to reduce medullary demand by inhibiting solute reabsorption and to attenuate the severity of acute renal injury in animal models. It is postulated that it may protect the human kidney from ischaemic injury. There are some small studies of low statistical power, but they are confounded by co-interventions such as low dose dopamine or mannitol. The best evidence to answer this current clinical question comes from the two observational cohorts identified in this review. The two studies collected over 2000 patients. They both document overall detriment with the use of diuretics with odds ratios of >1.0, as opposed to benefit, although the BEST investigators did not demonstrate statistical significance in their findings. The question of whether an RCT can be justified on the basis of these observational findings is debatable.

Clinical Bottom Line

In critically ill patients with acute renal failure, there is no evidence to suggest that the use of loop diuretics reduces mortality, reduces length of ITU/hospital stay or increases the recovery of renal function.

Level of Evidence

Level 2 - Studies considered were neither 1 or 3.

References

1. Uchino S. Doig GS. Bellomo R et al. Beginning and Ending Supportive Therapy for the Kidney (B.E.S.T. Kidney) Investigators. Diuretics and mortality in acute renal failure. Critical Care Medicine 2004;32(8):1669-77.
2. Mehta RL. Pascual MT. Soroko S et al. Diuretics, mortality, and nonrecovery of renal function in acute renal failure. JAMA. 2002;288(20):2547-53.